Security audit

Product Market Research

Security checks for vulnerabilities and agentic risk

Overview

This is a coherent product-market research skill, with disclosed web searching and an optional scheduled tracking feature users should enable carefully.

Before installing, expect this skill to perform external web searches, fetch pages, and sometimes ask you for links, screenshots, or files to analyze. Only enable the continuous tracking option if you are comfortable with a recurring cron job, and ask how to disable it before it is created.

A.I.G

Vulnerability Patterns

Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns

Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (6)

Vague Triggers

Medium

Confidence: 94% confidence
Finding: The trigger '任何涉及「产品 + 地区 + 市场信息收集」的请求' is a catch-all activation condition without clear boundaries or exclusions. This broad phrasing can overlap with common user requests and makes it unclear when the skill should activate versus when a general research workflow should handle the request.

Natural-Language Policy Violations

Medium

Confidence: 95% confidence
Finding: The instruction to translate non-Chinese pages into Chinese imposes a fixed language choice. The file does not offer the user a language preference or opt-in, so this is a language/locale policy issue rather than a justified region-specific constraint.

Natural-Language Policy Violations

Medium

Confidence: 97% confidence
Finding: These lines instruct the skill to process foreign-language material and '全部翻译为中文', which enforces a single output language. No user choice or opt-in is provided, so the file contains a natural-language language-policy violation.

Natural-Language Policy Violations

Medium

Confidence: 97% confidence
Finding: The document is entirely written in Chinese and presents mandatory process guidance without any indication that language selection is configurable or user-driven. In a general-purpose agent skill, forcing a single language can override user intent, reduce transparency for non-Chinese-speaking users, and cause analysts or downstream systems to miss critical instructions or mis-handle evidence weighting.

Natural-Language Policy Violations

Medium

Confidence: 93% confidence
Finding: The file is written as a prescriptive output template in Chinese and presents the format as the standard output format, which implies responses should be produced in Chinese. Because there is no opt-in or alternative language path, this can violate a language/locale policy requiring user choice rather than forced locale.

Natural-Language Policy Violations

Medium

Confidence: 93% confidence
Finding: The skill explicitly requires bilingual parallel searching using Chinese alongside the target region language without any user opt-in. This can unnecessarily disclose user interests, products, or locations to Chinese search providers or Chinese-language platforms, expanding data exposure and potentially conflicting with user expectations, privacy requirements, or jurisdictional constraints.

Static analysis

No suspicious patterns detected.